Contact Interface Verification for DYNA3D Scenario 1: Basic Contact

A suite of test problems has been developed to examine contact behavior within the nonlinear, three-dimensional, explicit finite element analysis (FEA) code DYNA3D (Lin, 2005). The test problems address the basic functionality of the contact algorithms, including the behavior of various kinematic, penalty, and Lagrangian enforcement formulations. The results from the DYNA3D analyses are compared to closed form solutions to verify the contact behavior. This work was performed as part of the Verification and Validation efforts of LLNL W Program within the NNSA's Advanced Simulation and Computing (ASC) Program. DYNA3D models the transient dynamic response of solids and structures including the interactions between disjoint bodies (parts). A wide variety of contact surfaces are available to represent the diverse interactions possible during an analysis, including relative motion (sliding), separation and gap closure (voids), and fixed relative position (tied). The problem geometry may be defined using a combination of element formulations, including one-dimensional beam and truss elements, two-dimensional shell elements, and three-dimensional solid elements. Consequently, it is necessary to consider various element interactions for each contact algorithm being verified. Most of the contact algorithms currently available in DYNA3D are examined; the exceptions are the Type 4--Single Surface Contact and Type 11--SAND algorithms. It is likely that these algorithms will be removed since their functionality is embodied in other, more robust, contact algorithms. The automatic contact algorithm is evaluated using the Type 12 interface. Two other variations of automatic contact, Type 13 and Type 14, offer additional means to adapt the interface domain, but share the same search and restoration algorithms as Type 12. The contact algorithms are summarized in Table 1. This report and associated test problems examine the scenario where one contact surface exists between two disjoint bodies. These test problems focus on whether a particular contact algorithm properly represents the interactions along the interface. A companion report (McMichael, 2006) and test problems address the multi-contact scenario in which multiple bodies interact with each other via multiple interfaces. The multi-contact test problems examine whether any ordering issues exist in the contact logic. The test problems are analyzed using version 5.2 (compiled on 12/22/2005) of DYNA3D. The analytical results are used to form baseline solutions for subsequent regression testing